[mirror_ubuntu-kernels.git] / fs / orangefs / inode.c

/*
 * (C) 2001 Clemson University and The University of Chicago
 *
 * See COPYING in top-level directory.
 */

/*
 *  Linux VFS inode operations.
 */

#include "protocol.h"
#include "orangefs-kernel.h"
#include "orangefs-bufmap.h"

static int read_one_page(struct page *page)
{
	int ret;
	int max_block;
	ssize_t bytes_read = 0;
	struct inode *inode = page->mapping->host;
	const __u32 blocksize = PAGE_CACHE_SIZE;	/* inode->i_blksize */
	const __u32 blockbits = PAGE_CACHE_SHIFT;	/* inode->i_blkbits */
	struct iov_iter to;
	struct bio_vec bv = {.bv_page = page, .bv_len = PAGE_SIZE};

	iov_iter_bvec(&to, ITER_BVEC | READ, &bv, 1, PAGE_SIZE);

	gossip_debug(GOSSIP_INODE_DEBUG,
		    "orangefs_readpage called with page %p\n",
		     page);

	max_block = ((inode->i_size / blocksize) + 1);

	if (page->index < max_block) {
		loff_t blockptr_offset = (((loff_t) page->index) << blockbits);

		bytes_read = orangefs_inode_read(inode,
						 &to,
						 &blockptr_offset,
						 inode->i_size);
	}
	/* this will only zero remaining unread portions of the page data */
	iov_iter_zero(~0U, &to);
	/* takes care of potential aliasing */
	flush_dcache_page(page);
	if (bytes_read < 0) {
		ret = bytes_read;
		SetPageError(page);
	} else {
		SetPageUptodate(page);
		if (PageError(page))
			ClearPageError(page);
		ret = 0;
	}
	/* unlock the page after the ->readpage() routine completes */
	unlock_page(page);
	return ret;
}

static int orangefs_readpage(struct file *file, struct page *page)
{
	return read_one_page(page);
}

static int orangefs_readpages(struct file *file,
			   struct address_space *mapping,
			   struct list_head *pages,
			   unsigned nr_pages)
{
	int page_idx;
	int ret;

	gossip_debug(GOSSIP_INODE_DEBUG, "orangefs_readpages called\n");

	for (page_idx = 0; page_idx < nr_pages; page_idx++) {
		struct page *page;

		page = list_entry(pages->prev, struct page, lru);
		list_del(&page->lru);
		if (!add_to_page_cache(page,
				       mapping,
				       page->index,
				       GFP_KERNEL)) {
			ret = read_one_page(page);
			gossip_debug(GOSSIP_INODE_DEBUG,
				"failure adding page to cache, read_one_page returned: %d\n",
				ret);
	      } else {
			page_cache_release(page);
	      }
	}
	BUG_ON(!list_empty(pages));
	return 0;
}

static void orangefs_invalidatepage(struct page *page,
				 unsigned int offset,
				 unsigned int length)
{
	gossip_debug(GOSSIP_INODE_DEBUG,
		     "orangefs_invalidatepage called on page %p "
		     "(offset is %u)\n",
		     page,
		     offset);

	ClearPageUptodate(page);
	ClearPageMappedToDisk(page);
	return;

}

static int orangefs_releasepage(struct page *page, gfp_t foo)
{
	gossip_debug(GOSSIP_INODE_DEBUG,
		     "orangefs_releasepage called on page %p\n",
		     page);
	return 0;
}

/*
 * Having a direct_IO entry point in the address_space_operations
 * struct causes the kernel to allows us to use O_DIRECT on
 * open. Nothing will ever call this thing, but in the future we
 * will need to be able to use O_DIRECT on open in order to support
 * AIO. Modeled after NFS, they do this too.
 */
/*
 * static ssize_t orangefs_direct_IO(int rw,
 *			struct kiocb *iocb,
 *			struct iov_iter *iter,
 *			loff_t offset)
 *{
 *	gossip_debug(GOSSIP_INODE_DEBUG,
 *		     "orangefs_direct_IO: %s\n",
 *		     iocb->ki_filp->f_path.dentry->d_name.name);
 *
 *	return -EINVAL;
 *}
 */

struct backing_dev_info orangefs_backing_dev_info = {
	.name = "orangefs",
	.ra_pages = 0,
	.capabilities = BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK,
};

/** ORANGEFS2 implementation of address space operations */
const struct address_space_operations orangefs_address_operations = {
	.readpage = orangefs_readpage,
	.readpages = orangefs_readpages,
	.invalidatepage = orangefs_invalidatepage,
	.releasepage = orangefs_releasepage,
/*	.direct_IO = orangefs_direct_IO */
};

static int orangefs_setattr_size(struct inode *inode, struct iattr *iattr)
{
	struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
	struct orangefs_kernel_op_s *new_op;
	loff_t orig_size = i_size_read(inode);
	int ret = -EINVAL;

	gossip_debug(GOSSIP_INODE_DEBUG,
		     "%s: %pU: Handle is %pU | fs_id %d | size is %llu\n",
		     __func__,
		     get_khandle_from_ino(inode),
		     &orangefs_inode->refn.khandle,
		     orangefs_inode->refn.fs_id,
		     iattr->ia_size);

	truncate_setsize(inode, iattr->ia_size);

	new_op = op_alloc(ORANGEFS_VFS_OP_TRUNCATE);
	if (!new_op)
		return -ENOMEM;

	new_op->upcall.req.truncate.refn = orangefs_inode->refn;
	new_op->upcall.req.truncate.size = (__s64) iattr->ia_size;

	ret = service_operation(new_op, __func__,
				get_interruptible_flag(inode));

	/*
	 * the truncate has no downcall members to retrieve, but
	 * the status value tells us if it went through ok or not
	 */
	gossip_debug(GOSSIP_INODE_DEBUG,
		     "orangefs: orangefs_truncate got return value of %d\n",
		     ret);

	op_release(new_op);

	if (ret != 0)
		return ret;

	/*
	 * Only change the c/mtime if we are changing the size or we are
	 * explicitly asked to change it.  This handles the semantic difference
	 * between truncate() and ftruncate() as implemented in the VFS.
	 *
	 * The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a
	 * special case where we need to update the times despite not having
	 * these flags set.  For all other operations the VFS set these flags
	 * explicitly if it wants a timestamp update.
	 */
	if (orig_size != i_size_read(inode) &&
	    !(iattr->ia_valid & (ATTR_CTIME | ATTR_MTIME))) {
		iattr->ia_ctime = iattr->ia_mtime =
			current_fs_time(inode->i_sb);
		iattr->ia_valid |= ATTR_CTIME | ATTR_MTIME;
	}

	return ret;
}

/*
 * Change attributes of an object referenced by dentry.
 */
int orangefs_setattr(struct dentry *dentry, struct iattr *iattr)
{
	int ret = -EINVAL;
	struct inode *inode = dentry->d_inode;

	gossip_debug(GOSSIP_INODE_DEBUG,
		     "orangefs_setattr: called on %s\n",
		     dentry->d_name.name);

	ret = inode_change_ok(inode, iattr);
	if (ret)
		goto out;

	if ((iattr->ia_valid & ATTR_SIZE) &&
	    iattr->ia_size != i_size_read(inode)) {
		ret = orangefs_setattr_size(inode, iattr);
		if (ret)
			goto out;
	}

	setattr_copy(inode, iattr);
	mark_inode_dirty(inode);

	ret = orangefs_inode_setattr(inode, iattr);
	gossip_debug(GOSSIP_INODE_DEBUG,
		     "orangefs_setattr: inode_setattr returned %d\n",
		     ret);

	if (!ret && (iattr->ia_valid & ATTR_MODE))
		/* change mod on a file that has ACLs */
		ret = posix_acl_chmod(inode, inode->i_mode);

out:
	gossip_debug(GOSSIP_INODE_DEBUG, "orangefs_setattr: returning %d\n", ret);
	return ret;
}

/*
 * Obtain attributes of an object given a dentry
 */
int orangefs_getattr(struct vfsmount *mnt,
		  struct dentry *dentry,
		  struct kstat *kstat)
{
	int ret = -ENOENT;
	struct inode *inode = dentry->d_inode;
	struct orangefs_inode_s *orangefs_inode = NULL;

	gossip_debug(GOSSIP_INODE_DEBUG,
		     "orangefs_getattr: called on %s\n",
		     dentry->d_name.name);

	ret = orangefs_inode_old_getattr(inode, ORANGEFS_ATTR_SYS_ALL_NOHINT,
	    0);
	if (ret == 0) {
		generic_fillattr(inode, kstat);

		/* override block size reported to stat */
		orangefs_inode = ORANGEFS_I(inode);
		kstat->blksize = orangefs_inode->blksize;

		inode->i_link = ORANGEFS_I(dentry->d_inode)->link_target;
	} else {
		/* assume an I/O error and flag inode as bad */
		gossip_debug(GOSSIP_INODE_DEBUG,
			     "%s:%s:%d calling make bad inode\n",
			     __FILE__,
			     __func__,
			     __LINE__);
		orangefs_make_bad_inode(inode);
	}
	return ret;
}

int orangefs_permission(struct inode *inode, int mask)
{
	int ret;

	if (mask & MAY_NOT_BLOCK)
		return -ECHILD;

	gossip_debug(GOSSIP_INODE_DEBUG, "%s: refreshing\n", __func__);

	/* Make sure the permission (and other common attrs) are up to date. */
	ret = orangefs_inode_old_getattr(inode,
	    ORANGEFS_ATTR_SYS_ALL_NOHINT_NOSIZE, 0);
	if (ret < 0)
		return ret;

	return generic_permission(inode, mask);
}

/* ORANGEDS2 implementation of VFS inode operations for files */
struct inode_operations orangefs_file_inode_operations = {
	.get_acl = orangefs_get_acl,
	.set_acl = orangefs_set_acl,
	.setattr = orangefs_setattr,
	.getattr = orangefs_getattr,
	.setxattr = generic_setxattr,
	.getxattr = generic_getxattr,
	.listxattr = orangefs_listxattr,
	.removexattr = generic_removexattr,
	.permission = orangefs_permission,
};

static int orangefs_init_iops(struct inode *inode)
{
	inode->i_mapping->a_ops = &orangefs_address_operations;

	switch (inode->i_mode & S_IFMT) {
	case S_IFREG:
		inode->i_op = &orangefs_file_inode_operations;
		inode->i_fop = &orangefs_file_operations;
		inode->i_blkbits = PAGE_CACHE_SHIFT;
		break;
	case S_IFLNK:
		inode->i_op = &orangefs_symlink_inode_operations;
		break;
	case S_IFDIR:
		inode->i_op = &orangefs_dir_inode_operations;
		inode->i_fop = &orangefs_dir_operations;
		break;
	default:
		gossip_debug(GOSSIP_INODE_DEBUG,
			     "%s: unsupported mode\n",
			     __func__);
		return -EINVAL;
	}

	return 0;
}

/*
 * Given a ORANGEFS object identifier (fsid, handle), convert it into a ino_t type
 * that will be used as a hash-index from where the handle will
 * be searched for in the VFS hash table of inodes.
 */
static inline ino_t orangefs_handle_hash(struct orangefs_object_kref *ref)
{
	if (!ref)
		return 0;
	return orangefs_khandle_to_ino(&(ref->khandle));
}

/*
 * Called to set up an inode from iget5_locked.
 */
static int orangefs_set_inode(struct inode *inode, void *data)
{
	struct orangefs_object_kref *ref = (struct orangefs_object_kref *) data;
	ORANGEFS_I(inode)->refn.fs_id = ref->fs_id;
	ORANGEFS_I(inode)->refn.khandle = ref->khandle;
	return 0;
}

/*
 * Called to determine if handles match.
 */
static int orangefs_test_inode(struct inode *inode, void *data)
{
	struct orangefs_object_kref *ref = (struct orangefs_object_kref *) data;
	struct orangefs_inode_s *orangefs_inode = NULL;

	orangefs_inode = ORANGEFS_I(inode);
	return (!ORANGEFS_khandle_cmp(&(orangefs_inode->refn.khandle), &(ref->khandle))
		&& orangefs_inode->refn.fs_id == ref->fs_id);
}

/*
 * Front-end to lookup the inode-cache maintained by the VFS using the ORANGEFS
 * file handle.
 *
 * @sb: the file system super block instance.
 * @ref: The ORANGEFS object for which we are trying to locate an inode structure.
 */
struct inode *orangefs_iget(struct super_block *sb, struct orangefs_object_kref *ref)
{
	struct inode *inode = NULL;
	unsigned long hash;
	int error;

	hash = orangefs_handle_hash(ref);
	inode = iget5_locked(sb, hash, orangefs_test_inode, orangefs_set_inode, ref);
	if (!inode || !(inode->i_state & I_NEW))
		return inode;

	error = orangefs_inode_getattr(inode, 1, 0);
	if (error) {
		iget_failed(inode);
		return ERR_PTR(error);
	}

	inode->i_ino = hash;	/* needed for stat etc */
	orangefs_init_iops(inode);
	unlock_new_inode(inode);

	gossip_debug(GOSSIP_INODE_DEBUG,
		     "iget handle %pU, fsid %d hash %ld i_ino %lu\n",
		     &ref->khandle,
		     ref->fs_id,
		     hash,
		     inode->i_ino);

	return inode;
}

/*
 * Allocate an inode for a newly created file and insert it into the inode hash.
 */
struct inode *orangefs_new_inode(struct super_block *sb, struct inode *dir,
		int mode, dev_t dev, struct orangefs_object_kref *ref)
{
	unsigned long hash = orangefs_handle_hash(ref);
	struct inode *inode;
	int error;

	gossip_debug(GOSSIP_INODE_DEBUG,
		     "%s:(sb is %p | MAJOR(dev)=%u | MINOR(dev)=%u mode=%o)\n",
		     __func__,
		     sb,
		     MAJOR(dev),
		     MINOR(dev),
		     mode);

	inode = new_inode(sb);
	if (!inode)
		return NULL;

	orangefs_set_inode(inode, ref);
	inode->i_ino = hash;	/* needed for stat etc */

	error = orangefs_inode_getattr(inode, 1, 0);
	if (error)
		goto out_iput;

	orangefs_init_iops(inode);

	inode->i_mode = mode;
	inode->i_uid = current_fsuid();
	inode->i_gid = current_fsgid();
	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	inode->i_size = PAGE_CACHE_SIZE;
	inode->i_rdev = dev;

	error = insert_inode_locked4(inode, hash, orangefs_test_inode, ref);
	if (error < 0)
		goto out_iput;

	gossip_debug(GOSSIP_INODE_DEBUG,
		     "Initializing ACL's for inode %pU\n",
		     get_khandle_from_ino(inode));
	orangefs_init_acl(inode, dir);
	return inode;

out_iput:
	iput(inode);
	return ERR_PTR(error);
}
Commit	Line	Data
5db11c21 MM	1	/*
	2	* (C) 2001 Clemson University and The University of Chicago
	3	*
	4	* See COPYING in top-level directory.
	5	*/
	6
	7	/*
	8	* Linux VFS inode operations.
	9	*/
	10
	11	#include "protocol.h"
575e9461 MM	12	#include "orangefs-kernel.h"
575e9461 MM	13	#include "orangefs-bufmap.h"
5db11c21 MM	14
	15	static int read_one_page(struct page *page)
	16	{
5db11c21 MM	17	int ret;
	18	int max_block;
	19	ssize_t bytes_read = 0;
	20	struct inode *inode = page->mapping->host;
	21	const __u32 blocksize = PAGE_CACHE_SIZE; /* inode->i_blksize */
	22	const __u32 blockbits = PAGE_CACHE_SHIFT; /* inode->i_blkbits */
74f68fce AV	23	struct iov_iter to;
	24	struct bio_vec bv = {.bv_page = page, .bv_len = PAGE_SIZE};
	25
	26	iov_iter_bvec(&to, ITER_BVEC \| READ, &bv, 1, PAGE_SIZE);
5db11c21 MM	27
5db11c21 MM	28	gossip_debug(GOSSIP_INODE_DEBUG,
8bb8aefd	29	"orangefs_readpage called with page %p\n",
5db11c21	30	page);
5db11c21 MM	31
	32	max_block = ((inode->i_size / blocksize) + 1);
	33
	34	if (page->index < max_block) {
	35	loff_t blockptr_offset = (((loff_t) page->index) << blockbits);
	36
8bb8aefd YL	37	bytes_read = orangefs_inode_read(inode,
	38	&to,
	39	&blockptr_offset,
	40	inode->i_size);
5db11c21	41	}
74f68fce AV	42	/* this will only zero remaining unread portions of the page data */
74f68fce AV	43	iov_iter_zero(~0U, &to);
5db11c21 MM	44	/* takes care of potential aliasing */
	45	flush_dcache_page(page);
	46	if (bytes_read < 0) {
	47	ret = bytes_read;
	48	SetPageError(page);
	49	} else {
	50	SetPageUptodate(page);
	51	if (PageError(page))
	52	ClearPageError(page);
	53	ret = 0;
	54	}
5db11c21 MM	55	/* unlock the page after the ->readpage() routine completes */
	56	unlock_page(page);
	57	return ret;
	58	}
	59
8bb8aefd	60	static int orangefs_readpage(struct file file, struct page page)
5db11c21 MM	61	{
	62	return read_one_page(page);
	63	}
	64
8bb8aefd	65	static int orangefs_readpages(struct file *file,
5db11c21 MM	66	struct address_space *mapping,
	67	struct list_head *pages,
	68	unsigned nr_pages)
	69	{
	70	int page_idx;
	71	int ret;
	72
8bb8aefd	73	gossip_debug(GOSSIP_INODE_DEBUG, "orangefs_readpages called\n");
5db11c21 MM	74
	75	for (page_idx = 0; page_idx < nr_pages; page_idx++) {
	76	struct page *page;
	77
	78	page = list_entry(pages->prev, struct page, lru);
	79	list_del(&page->lru);
	80	if (!add_to_page_cache(page,
	81	mapping,
	82	page->index,
	83	GFP_KERNEL)) {
	84	ret = read_one_page(page);
	85	gossip_debug(GOSSIP_INODE_DEBUG,
	86	"failure adding page to cache, read_one_page returned: %d\n",
	87	ret);
	88	} else {
	89	page_cache_release(page);
	90	}
	91	}
	92	BUG_ON(!list_empty(pages));
	93	return 0;
	94	}
	95
8bb8aefd	96	static void orangefs_invalidatepage(struct page *page,
5db11c21 MM	97	unsigned int offset,
	98	unsigned int length)
	99	{
	100	gossip_debug(GOSSIP_INODE_DEBUG,
8bb8aefd	101	"orangefs_invalidatepage called on page %p "
5db11c21 MM	102	"(offset is %u)\n",
	103	page,
	104	offset);
	105
	106	ClearPageUptodate(page);
	107	ClearPageMappedToDisk(page);
	108	return;
	109
	110	}
	111
8bb8aefd	112	static int orangefs_releasepage(struct page *page, gfp_t foo)
5db11c21 MM	113	{
5db11c21 MM	114	gossip_debug(GOSSIP_INODE_DEBUG,
8bb8aefd	115	"orangefs_releasepage called on page %p\n",
5db11c21 MM	116	page);
	117	return 0;
	118	}
	119
	120	/*
	121	* Having a direct_IO entry point in the address_space_operations
	122	* struct causes the kernel to allows us to use O_DIRECT on
	123	* open. Nothing will ever call this thing, but in the future we
	124	* will need to be able to use O_DIRECT on open in order to support
	125	* AIO. Modeled after NFS, they do this too.
	126	*/
	127	/*
575e9461	128	* static ssize_t orangefs_direct_IO(int rw,
54804949 MM	129	* struct kiocb *iocb,
	130	* struct iov_iter *iter,
	131	* loff_t offset)
	132	*{
	133	* gossip_debug(GOSSIP_INODE_DEBUG,
8bb8aefd	134	* "orangefs_direct_IO: %s\n",
54804949 MM	135	* iocb->ki_filp->f_path.dentry->d_name.name);
	136	*
	137	* return -EINVAL;
	138	*}
	139	*/
5db11c21	140
8bb8aefd YL	141	struct backing_dev_info orangefs_backing_dev_info = {
8bb8aefd YL	142	.name = "orangefs",
5db11c21 MM	143	.ra_pages = 0,
	144	.capabilities = BDI_CAP_NO_ACCT_DIRTY \| BDI_CAP_NO_WRITEBACK,
	145	};
	146
8bb8aefd YL	147	/** ORANGEFS2 implementation of address space operations */
	148	const struct address_space_operations orangefs_address_operations = {
	149	.readpage = orangefs_readpage,
	150	.readpages = orangefs_readpages,
	151	.invalidatepage = orangefs_invalidatepage,
	152	.releasepage = orangefs_releasepage,
575e9461	153	/* .direct_IO = orangefs_direct_IO */
5db11c21 MM	154	};
5db11c21 MM	155
8bb8aefd	156	static int orangefs_setattr_size(struct inode inode, struct iattr iattr)
5db11c21	157	{
8bb8aefd YL	158	struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
8bb8aefd YL	159	struct orangefs_kernel_op_s *new_op;
5db11c21 MM	160	loff_t orig_size = i_size_read(inode);
	161	int ret = -EINVAL;
	162
	163	gossip_debug(GOSSIP_INODE_DEBUG,
	164	"%s: %pU: Handle is %pU \| fs_id %d \| size is %llu\n",
	165	__func__,
	166	get_khandle_from_ino(inode),
8bb8aefd YL	167	&orangefs_inode->refn.khandle,
8bb8aefd YL	168	orangefs_inode->refn.fs_id,
5db11c21 MM	169	iattr->ia_size);
	170
	171	truncate_setsize(inode, iattr->ia_size);
	172
8bb8aefd	173	new_op = op_alloc(ORANGEFS_VFS_OP_TRUNCATE);
5db11c21 MM	174	if (!new_op)
	175	return -ENOMEM;
	176
8bb8aefd	177	new_op->upcall.req.truncate.refn = orangefs_inode->refn;
5db11c21 MM	178	new_op->upcall.req.truncate.size = (__s64) iattr->ia_size;
	179
	180	ret = service_operation(new_op, __func__,
	181	get_interruptible_flag(inode));
	182
	183	/*
	184	* the truncate has no downcall members to retrieve, but
	185	* the status value tells us if it went through ok or not
	186	*/
	187	gossip_debug(GOSSIP_INODE_DEBUG,
8bb8aefd	188	"orangefs: orangefs_truncate got return value of %d\n",
5db11c21 MM	189	ret);
	190
	191	op_release(new_op);
	192
	193	if (ret != 0)
	194	return ret;
	195
	196	/*
	197	* Only change the c/mtime if we are changing the size or we are
	198	* explicitly asked to change it. This handles the semantic difference
	199	* between truncate() and ftruncate() as implemented in the VFS.
	200	*
	201	* The regular truncate() case without ATTR_CTIME and ATTR_MTIME is a
	202	* special case where we need to update the times despite not having
	203	* these flags set. For all other operations the VFS set these flags
	204	* explicitly if it wants a timestamp update.
	205	*/
	206	if (orig_size != i_size_read(inode) &&
	207	!(iattr->ia_valid & (ATTR_CTIME \| ATTR_MTIME))) {
	208	iattr->ia_ctime = iattr->ia_mtime =
	209	current_fs_time(inode->i_sb);
	210	iattr->ia_valid \|= ATTR_CTIME \| ATTR_MTIME;
	211	}
	212
	213	return ret;
	214	}
	215
	216	/*
	217	* Change attributes of an object referenced by dentry.
	218	*/
8bb8aefd	219	int orangefs_setattr(struct dentry dentry, struct iattr iattr)
5db11c21 MM	220	{
	221	int ret = -EINVAL;
	222	struct inode *inode = dentry->d_inode;
	223
	224	gossip_debug(GOSSIP_INODE_DEBUG,
8bb8aefd	225	"orangefs_setattr: called on %s\n",
5db11c21 MM	226	dentry->d_name.name);
	227
	228	ret = inode_change_ok(inode, iattr);
	229	if (ret)
	230	goto out;
	231
	232	if ((iattr->ia_valid & ATTR_SIZE) &&
	233	iattr->ia_size != i_size_read(inode)) {
8bb8aefd	234	ret = orangefs_setattr_size(inode, iattr);
5db11c21 MM	235	if (ret)
	236	goto out;
	237	}
	238
	239	setattr_copy(inode, iattr);
	240	mark_inode_dirty(inode);
	241
8bb8aefd	242	ret = orangefs_inode_setattr(inode, iattr);
5db11c21	243	gossip_debug(GOSSIP_INODE_DEBUG,
8bb8aefd	244	"orangefs_setattr: inode_setattr returned %d\n",
5db11c21 MM	245	ret);
	246
	247	if (!ret && (iattr->ia_valid & ATTR_MODE))
	248	/* change mod on a file that has ACLs */
	249	ret = posix_acl_chmod(inode, inode->i_mode);
	250
	251	out:
8bb8aefd	252	gossip_debug(GOSSIP_INODE_DEBUG, "orangefs_setattr: returning %d\n", ret);
5db11c21 MM	253	return ret;
	254	}
	255
	256	/*
	257	* Obtain attributes of an object given a dentry
	258	*/
8bb8aefd	259	int orangefs_getattr(struct vfsmount *mnt,
5db11c21 MM	260	struct dentry *dentry,
	261	struct kstat *kstat)
	262	{
	263	int ret = -ENOENT;
	264	struct inode *inode = dentry->d_inode;
8bb8aefd	265	struct orangefs_inode_s *orangefs_inode = NULL;
5db11c21 MM	266
5db11c21 MM	267	gossip_debug(GOSSIP_INODE_DEBUG,
8bb8aefd	268	"orangefs_getattr: called on %s\n",
5db11c21 MM	269	dentry->d_name.name);
5db11c21 MM	270
3c9cf98d MB	271	ret = orangefs_inode_old_getattr(inode, ORANGEFS_ATTR_SYS_ALL_NOHINT,
3c9cf98d MB	272	0);
5db11c21 MM	273	if (ret == 0) {
5db11c21 MM	274	generic_fillattr(inode, kstat);
a7d3e78a	275
5db11c21	276	/* override block size reported to stat */
8bb8aefd YL	277	orangefs_inode = ORANGEFS_I(inode);
8bb8aefd YL	278	kstat->blksize = orangefs_inode->blksize;
a7d3e78a MM	279
a7d3e78a MM	280	inode->i_link = ORANGEFS_I(dentry->d_inode)->link_target;
5db11c21 MM	281	} else {
	282	/* assume an I/O error and flag inode as bad */
	283	gossip_debug(GOSSIP_INODE_DEBUG,
	284	"%s:%s:%d calling make bad inode\n",
	285	__FILE__,
	286	__func__,
	287	__LINE__);
8bb8aefd	288	orangefs_make_bad_inode(inode);
5db11c21 MM	289	}
	290	return ret;
	291	}
	292
933287da MB	293	int orangefs_permission(struct inode *inode, int mask)
	294	{
	295	int ret;
	296
	297	if (mask & MAY_NOT_BLOCK)
	298	return -ECHILD;
	299
	300	gossip_debug(GOSSIP_INODE_DEBUG, "%s: refreshing\n", __func__);
	301
	302	/* Make sure the permission (and other common attrs) are up to date. */
3c9cf98d	303	ret = orangefs_inode_old_getattr(inode,
933287da MB	304	ORANGEFS_ATTR_SYS_ALL_NOHINT_NOSIZE, 0);
	305	if (ret < 0)
	306	return ret;
	307
	308	return generic_permission(inode, mask);
	309	}
	310
8bb8aefd YL	311	/* ORANGEDS2 implementation of VFS inode operations for files */
	312	struct inode_operations orangefs_file_inode_operations = {
	313	.get_acl = orangefs_get_acl,
	314	.set_acl = orangefs_set_acl,
	315	.setattr = orangefs_setattr,
	316	.getattr = orangefs_getattr,
5db11c21 MM	317	.setxattr = generic_setxattr,
5db11c21 MM	318	.getxattr = generic_getxattr,
8bb8aefd	319	.listxattr = orangefs_listxattr,
5db11c21	320	.removexattr = generic_removexattr,
933287da	321	.permission = orangefs_permission,
5db11c21 MM	322	};
5db11c21 MM	323
8bb8aefd	324	static int orangefs_init_iops(struct inode *inode)
5db11c21	325	{
8bb8aefd	326	inode->i_mapping->a_ops = &orangefs_address_operations;
5db11c21 MM	327
	328	switch (inode->i_mode & S_IFMT) {
	329	case S_IFREG:
8bb8aefd YL	330	inode->i_op = &orangefs_file_inode_operations;
8bb8aefd YL	331	inode->i_fop = &orangefs_file_operations;
5db11c21 MM	332	inode->i_blkbits = PAGE_CACHE_SHIFT;
	333	break;
	334	case S_IFLNK:
8bb8aefd	335	inode->i_op = &orangefs_symlink_inode_operations;
5db11c21 MM	336	break;
5db11c21 MM	337	case S_IFDIR:
8bb8aefd YL	338	inode->i_op = &orangefs_dir_inode_operations;
8bb8aefd YL	339	inode->i_fop = &orangefs_dir_operations;
5db11c21 MM	340	break;
	341	default:
	342	gossip_debug(GOSSIP_INODE_DEBUG,
	343	"%s: unsupported mode\n",
	344	__func__);
	345	return -EINVAL;
	346	}
	347
	348	return 0;
	349	}
	350
	351	/*
8bb8aefd	352	* Given a ORANGEFS object identifier (fsid, handle), convert it into a ino_t type
5db11c21 MM	353	* that will be used as a hash-index from where the handle will
	354	* be searched for in the VFS hash table of inodes.
	355	*/
8bb8aefd	356	static inline ino_t orangefs_handle_hash(struct orangefs_object_kref *ref)
5db11c21 MM	357	{
	358	if (!ref)
	359	return 0;
8bb8aefd	360	return orangefs_khandle_to_ino(&(ref->khandle));
5db11c21 MM	361	}
	362
	363	/*
	364	* Called to set up an inode from iget5_locked.
	365	*/
8bb8aefd	366	static int orangefs_set_inode(struct inode inode, void data)
5db11c21	367	{
8bb8aefd	368	struct orangefs_object_kref ref = (struct orangefs_object_kref ) data;
a4c680a0 MB	369	ORANGEFS_I(inode)->refn.fs_id = ref->fs_id;
a4c680a0 MB	370	ORANGEFS_I(inode)->refn.khandle = ref->khandle;
5db11c21 MM	371	return 0;
	372	}
	373
	374	/*
	375	* Called to determine if handles match.
	376	*/
8bb8aefd	377	static int orangefs_test_inode(struct inode inode, void data)
5db11c21	378	{
8bb8aefd YL	379	struct orangefs_object_kref ref = (struct orangefs_object_kref ) data;
8bb8aefd YL	380	struct orangefs_inode_s *orangefs_inode = NULL;
5db11c21	381
8bb8aefd YL	382	orangefs_inode = ORANGEFS_I(inode);
	383	return (!ORANGEFS_khandle_cmp(&(orangefs_inode->refn.khandle), &(ref->khandle))
	384	&& orangefs_inode->refn.fs_id == ref->fs_id);
5db11c21 MM	385	}
	386
	387	/*
8bb8aefd	388	* Front-end to lookup the inode-cache maintained by the VFS using the ORANGEFS
5db11c21 MM	389	* file handle.
	390	*
	391	* @sb: the file system super block instance.
8bb8aefd	392	* @ref: The ORANGEFS object for which we are trying to locate an inode structure.
5db11c21	393	*/
8bb8aefd	394	struct inode orangefs_iget(struct super_block sb, struct orangefs_object_kref *ref)
5db11c21 MM	395	{
	396	struct inode *inode = NULL;
	397	unsigned long hash;
	398	int error;
	399
8bb8aefd YL	400	hash = orangefs_handle_hash(ref);
8bb8aefd YL	401	inode = iget5_locked(sb, hash, orangefs_test_inode, orangefs_set_inode, ref);
5db11c21 MM	402	if (!inode \|\| !(inode->i_state & I_NEW))
	403	return inode;
	404
075cca50	405	error = orangefs_inode_getattr(inode, 1, 0);
5db11c21 MM	406	if (error) {
	407	iget_failed(inode);
	408	return ERR_PTR(error);
	409	}
	410
	411	inode->i_ino = hash; /* needed for stat etc */
8bb8aefd	412	orangefs_init_iops(inode);
5db11c21 MM	413	unlock_new_inode(inode);
	414
	415	gossip_debug(GOSSIP_INODE_DEBUG,
	416	"iget handle %pU, fsid %d hash %ld i_ino %lu\n",
	417	&ref->khandle,
	418	ref->fs_id,
	419	hash,
	420	inode->i_ino);
	421
	422	return inode;
	423	}
	424
	425	/*
	426	* Allocate an inode for a newly created file and insert it into the inode hash.
	427	*/
8bb8aefd YL	428	struct inode orangefs_new_inode(struct super_block sb, struct inode *dir,
8bb8aefd YL	429	int mode, dev_t dev, struct orangefs_object_kref *ref)
5db11c21	430	{
8bb8aefd	431	unsigned long hash = orangefs_handle_hash(ref);
5db11c21 MM	432	struct inode *inode;
	433	int error;
	434
	435	gossip_debug(GOSSIP_INODE_DEBUG,
5253487e MM	436	"%s:(sb is %p \| MAJOR(dev)=%u \| MINOR(dev)=%u mode=%o)\n",
5253487e MM	437	__func__,
5db11c21 MM	438	sb,
	439	MAJOR(dev),
	440	MINOR(dev),
	441	mode);
	442
	443	inode = new_inode(sb);
	444	if (!inode)
	445	return NULL;
	446
8bb8aefd	447	orangefs_set_inode(inode, ref);
5db11c21 MM	448	inode->i_ino = hash; /* needed for stat etc */
5db11c21 MM	449
075cca50	450	error = orangefs_inode_getattr(inode, 1, 0);
5db11c21 MM	451	if (error)
	452	goto out_iput;
	453
8bb8aefd	454	orangefs_init_iops(inode);
5db11c21 MM	455
	456	inode->i_mode = mode;
	457	inode->i_uid = current_fsuid();
	458	inode->i_gid = current_fsgid();
	459	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
	460	inode->i_size = PAGE_CACHE_SIZE;
	461	inode->i_rdev = dev;
	462
8bb8aefd	463	error = insert_inode_locked4(inode, hash, orangefs_test_inode, ref);
5db11c21 MM	464	if (error < 0)
	465	goto out_iput;
	466
	467	gossip_debug(GOSSIP_INODE_DEBUG,
	468	"Initializing ACL's for inode %pU\n",
	469	get_khandle_from_ino(inode));
8bb8aefd	470	orangefs_init_acl(inode, dir);
5db11c21 MM	471	return inode;
	472
	473	out_iput:
	474	iput(inode);
	475	return ERR_PTR(error);
	476	}